Choosing donors in clinical practice

In this issue of Blood, Montoro et al report higher overall survival after transplantation of grafts from human leukocyte antigen (HLA)–matched unrelated donors compared with HLA-mismatched unrelated and haploidentical donors for severe aplastic anemia in patients aged ≤40 years.¹ 

The conventional treatment algorithm for acquired severe aplastic anemia prioritizes transplantation of bone marrow from an HLA-matched sibling for patients aged ≤40 years; and for those without an HLA-matched sibling, immunosuppressive therapy.² Alternative donor transplantation is reserved for those who fail immunosuppressive therapy.² The importance of donor-recipient HLA match for unrelated donor transplantation and the challenges of identifying an HLA-matched unrelated donor for persons other than Whites has been reported by others.^3,4 In brief, unrelated donor selection should be limited to an HLA-matched or 1 HLA-locus mismatched unrelated donor to minimize graft failure and mortality risks.³ Overcoming the histocompatibility barrier associated with haploidentical donor transplantation using posttransplant cyclophosphamide has lowered the risks for graft failure and graft-versus-host disease and extended survival for patients with severe aplastic anemia without an HLA-matched or mismatched unrelated donor.⁵ Identifying an optimal alternative donor should ideally be studied in a prospective randomized trial but there are none given the rarity of the disease, complexities of trial design, and the time it would take to complete such a trial. Therefore, Montoro et al studied outcomes after alternative donor transplantation, retrospectively, using data reported to the European Group for Blood and Marrow Transplantation. Their study adds value to the existing literature through demonstration, in a relatively large cohort of transplantations for severe aplastic anemia, that an HLA-matched unrelated donor is preferred when such a donor is available. For those without an HLA-matched unrelated donor, either an HLA-mismatched unrelated or haploidentical donor is an acceptable choice despite higher risks for graft failure and mortality. The study failed to identify appreciable differences in outcomes after HLA-mismatched unrelated and haploidentical donor transplants. These findings extend our knowledge on outcomes of alternative donor transplantation in severe aplastic anemia.

Unrelated donor transplantation in severe aplastic anemia has been practiced for several decades, and modifications to transplant conditioning regimen, graft-versus-host disease prophylaxis, and graft type have ensured continued success.^6,7 On the other hand, the practice of using haploidentical donors for severe aplastic anemia transplants is relatively recent.⁵ Although Montoro et al limited their haploidentical transplants to those that included posttransplant cyclophosphamide for graft-versus-host disease prophylaxis, transplant conditioning regimens, serotherapy, and graft varied and may explain the higher graft failure and lower survival.¹ DeZern et al, at the Johns Hopkins Hospital, have adopted an approach that includes rabbit antithymocyte globulin (4.5 mg/kg), fludarabine (150 mg/m²), cyclophosphamide (29 mg/kg), single-fraction total body irradiation (200 cGy), infusion of bone marrow graft, and graft-versus-host disease prophylaxis that included cyclophosphamide posttransplant (100 mg/kg), mycophenolate mofetil, and tacrolimus and was recently studied in a multicenter phase 2 trial in the United States.⁸ The results of that trial⁸ showed that 16% (5 of 31) of patients developed graft failure, and the 1-year survival was 81%. Although the trial⁸ recommended a target yield of 4 × 10⁸ nucleated marrow cells per kilogram of recipient ideal body weight and a minimum yield of 2.5 × 10⁸ nucleated marrow cells per kilogram of recipient ideal body weight, 4 patients did not receive the minimum marrow counts. Three of the 4 patients developed graft failure, underscoring the need for adequate nucleated marrow cells in addition to absence of donor-specific antibodies (at mean fluorescence intensity >1000 by solid phase immunoassay).⁸ Longer follow-up of these patients is needed to confirm the excellent 1-year survival reported in that trial.⁸ 

An interesting observation in the study by Montoro et al is the upper age limit of 40 years at transplantation. The natural history of aplastic anemia has a bimodal distribution, with the second peak at ages >60 years.⁹ Yet the data reported to the European Transplant registry did not include anyone aged >40 years and presumably reflect practice variation between Europe and the United States. Although the median age at transplantation in the United States is comparable to that reported from Europe, 20% to 25% of patients undergoing alternative donor transplantation are aged >40 years at transplantation.^7,8 

When choosing alternative donors, an HLA-matched unrelated donor is ideal if such donor is available. For those without an HLA-matched donor, there are several questions regarding the most suitable HLA-mismatched alternative donor and how we may improve on the reported outcomes after HLA-mismatched and haploidentical donor transplants. Although Montoro et al did not observe difference between the 2 HLA-mismatched alternative donors, this may differ when conditioning regimen and graft-versus-host disease prophylaxis are optimized for both donor types. The Johns Hopkins approach for haploidentical transplants showed 1-year survival was 81%, a substantial improvement on the 63% after haploidentical transplants that was observed in the registry study.^1,8 To test the hypothesis that haploidentical donor transplants have better outcomes compared with HLA-mismatched unrelated donor transplants, we may consider a randomized trial of the 2 donor types in severe aplastic anemia. An approach that incorporates an optimized transplant conditioning regimen and graft-versus-host disease prophylaxis for each of the donor types coupled with a composite end point, such as graft failure–free survival or graft-versus-host disease and relapse-free survival, may identify which of the 2 HLA-mismatched donor types are preferred when an HLA-matched unrelated donor is not available.

Conflict-of-interest disclosure: The author declares no competing financial interests.